CN117031521A - Elastic fusion positioning method and system in indoor and outdoor seamless environment - Google Patents

Abstract

The invention relates to the technical field of positioning, and discloses an elastic fusion positioning method and system in an indoor and outdoor seamless environment, wherein the method comprises the following steps: inputting the GNSS observation vector and the SINS estimated observation vector into a first sub-filter for differencing, and updating measurement information by combining measurement noise covariance matrix and information distributed by a main filter; inputting the base station observation vector and the SINS calculated observation vector into a second sub-filter for differencing, and updating measurement information by combining measurement noise covariance matrix and information distributed by a main filter; and carrying out information fusion on updated measurement information of the first sub-filter and the second sub-filter in the main filter to obtain a final positioning result, and distributing the fusion information to the first sub-filter and the second sub-filter according to an information distribution criterion. The sensor fault and the non-line-of-sight error are effectively isolated, and the positioning accuracy is improved under the condition that the satellite part is unlocked in the indoor and outdoor transition areas.

Description

An elastic fusion positioning method and system in indoor and outdoor seamless environments

Technical field

The present invention relates to the field of positioning technology. Specifically, it relates to an elastic fusion positioning method and system in an indoor and outdoor seamless environment.

Background technique

The statements in this section merely provide background technical information related to the present invention and do not necessarily constitute prior art.

With the continuous improvement of urbanization and underground space development, location services need to meet the needs of high-precision and robust positioning in different indoor and outdoor scenarios. For navigation systems, the use of a single sensor often cannot meet the performance requirements for seamless indoor and outdoor positioning.

The Global Navigation Satellite System (GNSS) uses receivers to simultaneously measure the distances of multiple satellites to achieve high-precision calculation of the carrier position and speed; limited by the ground power level, its signals are susceptible to indoor and urban complexities. Environmental occlusion prevents normal operation.

Ultra-wideband technology calculates the distance between the two by recording the two-way propagation time of the pulse signal from the mobile station to the base station. It requires the existence of at least 4 visible base stations for effective positioning; this method is susceptible to the influence of complex indoor and outdoor observation environments, resulting in The number of base stations within the visual range of the mobile station is less than 4, which affects the positioning success rate. Moreover, ultra-wideband technology has limited coverage and cannot achieve wide-area positioning.

By loosely combining the above equipment with the strap-down inertial navigation system (SINS) position domain, when GNSS or ultra-wideband cannot work properly, navigation results can be obtained by relying on pure SINS recursion; the disadvantage is that when satellites are visible When the number of base stations is less than 4, because GNSS or ultra-wideband cannot be positioned individually and cannot be combined, possible observation information is wasted, and the SINS position error will gradually diverge.

Tightly combining GNSS, ultra-wideband and SINS in the measurement domain can effectively utilize the remaining observation information. However, under extreme conditions, such as only one base station being visible, the distance observations generated at this time are not enough to provide constraints for SINS. , resulting in a large positioning error. The 5G base station obtains the signal Time of Arrival (TOA) and Angle of Arrival (AOA) information through the channel parameter estimation algorithm, which alleviates the problem of insufficient observation values. In theory, a single base station can perform positioning, and through SINS The combination can further reduce the impact of harsh observation environments on 5G positioning; however, measurement accuracy is still affected by the environment. Existing methods lack evaluation of the availability of observation values and cannot effectively judge Non Line of Sight (NLOS) errors.

Contents of the invention

In order to solve the above problems, the present invention provides an elastic fusion positioning method and system in indoor and outdoor seamless environments, which effectively isolates sensor faults and non-line-of-sight errors. In the indoor and outdoor transition areas, the satellite is partially out of lock. , improved positioning accuracy compared to.

In order to achieve the above objects, the present invention adopts the following technical solutions:

The first aspect of the present invention provides an elastic fusion positioning method in indoor and outdoor seamless environments, which includes:

Obtain the GNSS observation vector, base station observation vector and SINS calculated observation vector;

Based on the GNSS observation vector and the base station observation vector, update the measurement noise covariance matrix;

The GNSS observation vector and the SINS-derived observation vector are input into the first sub-filter for difference, and the measurement information is updated based on the measurement noise covariance matrix and the information allocated by the main filter; at the same time, the base station observation vector is combined with the The observation vector calculated by SINS is input into the second sub-filter for difference, and combined with the information of the measurement noise covariance matrix and main filter allocation, the measurement information is updated;

The updated measurement information of the first sub-filter and the second sub-filter is fused in the main filter to obtain the final positioning result, and the fused information is distributed to the first sub-filter and the second sub-filter according to the information allocation criteria. Two sub-filters.

Further, the GNSS observation vector includes pseudorange and carrier phase.

Further, the base station observation vector includes the azimuth angle, altitude angle and signal propagation time of the terminal relative to the base station.

Further, the step of updating the measurement noise covariance matrix is:

Obtain the observation innovation vector, combine it with the judgment threshold, and calculate the variance expansion factor;

After vector diagonalizing the variance expansion factor, the measurement noise covariance matrix is updated.

Further, the state error covariance matrix in the fusion information is the reciprocal of the sum of the state error covariance matrices obtained by the two sub-filters;

Alternatively, the noise covariance matrix in the fusion information is the reciprocal of the sum of the process noise covariance matrices obtained by the two sub-filters.

Further, the global optimal estimate in the fusion information is related to the state error covariance matrix obtained by the two sub-filters, the local optimal solution obtained by the two sub-filters and the state error covariance matrix in the fusion information.

Further, the information distribution coefficient in the information distribution criterion is adjusted according to the spatial geometric intensity factor and carrier-to-noise ratio of the satellite distribution of GNSS.

The second aspect of the present invention provides an elastic fusion positioning system in indoor and outdoor seamless environments, which includes:

A data acquisition module configured to: acquire GNSS observation vectors, base station observation vectors, and SINS-derived observation vectors;

A fault detection module configured to: update the measurement noise covariance matrix based on the GNSS observation vector and the base station observation vector;

The measurement information update module is configured to: input the GNSS observation vector and the SINS estimated observation vector into the first sub-filter for difference, and combine the measurement noise covariance matrix and the information allocated by the main filter to perform measurement The measurement information is updated; at the same time, the base station observation vector and the SINS-derived observation vector are input into the second sub-filter for difference, and the measurement information is updated based on the measurement noise covariance matrix and the information allocated by the main filter;

A positioning module configured to: fuse the updated measurement information of the first sub-filter and the second sub-filter in the main filter to obtain the final positioning result, and distribute the fused information according to the information distribution criteria. to the first sub-filter and the second sub-filter.

Further, the GNSS observation vector includes pseudorange and carrier phase.

Further, the base station observation vector includes the azimuth angle, altitude angle and signal propagation time of the terminal relative to the base station.

Compared with the prior art, the beneficial effects of the present invention are:

The present invention provides an elastic fusion positioning method in indoor and outdoor seamless environments, which realizes tight combination of observation value levels in sub-filters, while designing fault detection and processing modules to suppress the influence of abnormal observation values, and finally in the main filter The sensor dynamically fuses and allocates information based on the quality of observation values, which can effectively isolate sensor faults and non-line-of-sight errors. In indoor and outdoor transition areas and under the condition that the satellite is partially out of lock, the positioning accuracy is improved by 68% compared to the GNSS-SINS tight combination. , the effective range is larger than the combined navigation algorithm based on GNSS, ultra-wideband and SINS, it can work continuously and stably in indoor and outdoor seamless environments, and the positioning accuracy is higher.

Description of the drawings

The accompanying drawings, which constitute part of the present invention, are used to provide a further understanding of the present invention. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute a limitation of the present invention.

Figure 1 is a flow chart of a flexible fusion positioning method in an indoor and outdoor seamless environment according to Embodiment 1 of the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and examples.

It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

If there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will be further described below with reference to the drawings and embodiments.

Embodiment 1

The purpose of the first embodiment is to provide a flexible fusion positioning method in indoor and outdoor seamless environments.

In order to solve the problem of signal blocking in indoor and outdoor seamless environments, this embodiment provides a flexible fusion positioning method in indoor and outdoor seamless environments, which integrates three sensors: GNSS, 5G base station and SINS.

This embodiment provides a flexible fusion positioning method in a seamless indoor and outdoor environment. In order to solve the problem of frequent loss of satellite lock and non-line-of-sight errors in a seamless indoor and outdoor environment, the sub-filter (GNSS-SINS filter and 5G-SINS filter) to achieve tight combination at the observation level, while designing fault detection and processing modules to suppress the influence of abnormal observation values. Finally, the information of each sub-filter is fused in the main filter, and based on the quality of the observation values Perform dynamic information distribution.

This embodiment provides a flexible fusion positioning method in indoor and outdoor seamless environments. In indoor and outdoor seamless environments, GNSS, 5G base stations and SINS are combined to design fault detection and processing, as well as elastic federated filtering, to achieve Integrated navigation of land vehicles belongs to the category of land integrated navigation.

This embodiment provides a flexible fusion positioning method in indoor and outdoor seamless environments, as shown in Figure 1, including the following steps:

Step 1. Obtain GNSS observation values and 5G base station observation values.

Among them, GNSS observation values include pseudorange and carrier phase. Specifically, GNSS observations can be provided by RTK (Real-timekinematic, real-time dynamic) mode, and pseudorange and carrier phase observations that eliminate errors in the ionosphere and troposphere are obtained through the GNSS reference station.

Among them, the 5G base station observation values include: the azimuth angle, altitude angle, signal propagation time, and signal propagation distance of the terminal (5G signal transmitting antenna mounted on the land carrier) relative to the 5G base station. The 5G base station receives the co-band Positioning Reference Signal (PRS) and can obtain the azimuth angle, altitude angle and signal propagation time of the terminal relative to the 5G base station. Taking the propagation speed of the signal as the speed of light, the signal propagation distance can be calculated.

Step 2. For GNSS observation values and 5G base station observation values, perform fault detection and processing respectively, and update the measurement noise covariance matrix.

GNSS observations and 5G base station observations in indoor and outdoor seamless environments are easily affected by NLOS errors, which in turn affects positioning accuracy.

Let the module of the observed innovation vector δY _i,k as a test value and use Distribution constructs fault detection thresholds to identify NLOS errors. Among them, δY _i,k is the observation innovation vector of the i- th sub-filter at time k , i is 1 or 2, when i is 1, the sub-filter is a GNSS-SINS filter, and when i is 2, the sub-filter is a GNSS-SINS filter. The filter is the 5G-SINS filter; Y _i,k is the observation vector (i.e. GNSS observation vector or 5G base station observation vector) input to the i- th sub-filter at time k ; H _i,k is the i-th sub-filter at time k The measurement relationship matrix;/> is the predicted state vector of the i-th sub-filter at time k .

When the measurement information is normal, the observation innovation vector δY _i,k is Gaussian white noise and obeys the normal distribution with a mean value of 0; when a fault occurs, that is, there is a gross error in the measurement information, then the mean value of δY _i,k is no longer 0, the fault can be determined through hypothesis testing.

Construct the hypothesis test statistic T _i,k as:

(1)

In the formula, is the covariance matrix of the observed innovation vector; T _i,k obeys/> with degrees of freedom n ( n is the dimension of the observation vector) Distribution, that is/> ;/> is the covariance matrix of the observation innovation vector, and R _i,k is the measurement noise covariance matrix.

Taking the significance level as α , the threshold for judgment of failure is:

(2)

When Ti _,k ≤ T _D , it is determined that there is no abnormality currently, otherwise it is determined that the current epoch observation vector (GNSS observation vector or 5G base station observation vector) contains an error.

The observation vector error is weakened by constructing the variance expansion factor α _i,k , that is, the observation innovation vector is obtained, combined with the judgment threshold, the variance expansion factor is calculated:

(3)

Diagonalize the α _i,k vector in the above formula to obtain diag ( α _i,k ), then modify the measurement noise covariance matrix as:

(4)

Will As the new measurement noise covariance matrix is substituted into Equation (20), the weight of the observation vector in the measurement update can be adaptively adjusted according to the quality of the observation vector, thereby making the filtering results more stable and more accurate. high.

Step 3. Update the observation vector calculated by SINS.

According to the current epoch SINS measurement update position (obtained by SINS mechanical programming) and GNSS ephemeris, after correction by the lever arm, the pseudorange and carrier phase from the satellite to the receiver antenna are obtained by back-reduction.

According to the current epoch SINS measurement update position and the known 5G base station coordinates, after correction by the lever arm, the azimuth angle, altitude angle and signal propagation time from the 5G base station to the terminal are obtained by back-reduction.

Step 4. Input the GNSS observation vector obtained in step 1 and the observation vector calculated by SINS in step 3 into the GNSS-SINS filter (first sub-filter). Combine the 5G base station observation vector obtained in step 1 with the SINS calculation in step 3. The observation vectors are input together into the 5G-SINS filter (second sub-filter).

(1) Construct the navigation state equation.

The establishment of the navigation state equation is related to the inertial sensor error type, combination method and the choice of coordinate system. For multi-source sensors, fusion of the common parts of different state vectors should also be considered.

First, construct a 15-dimensional state vector under the Earth Centered, Earth Fixed, ECEF system:

(5)

In the formula, X _t is the state vector at time t ; is the attitude angle error along the X, Y and Z directions under ECEF;/> is the velocity error along the X, Y and Z directions under ECEF;/> are the position errors in the X, Y and Z directions under ECEF; b _a and b _g are the accelerometer bias errors and gyroscope bias errors along the X, Y and Z directions in the carrier coordinate system, respectively.

Secondly, the navigation state equation is:

(6)

In the formula, F _t is the state transition matrix; is the process noise vector, including the random white noise w _ra of the accelerometer, the random white noise w _rg of the gyroscope, and the bias random walk process noise w _bad and w _bgd .

The state transition matrix F _t can be expressed as:

(7)

Among them, τ _s is the update time interval; is the skew symmetric matrix of the earth's rotation angular velocity;/> ,/> is the antisymmetric matrix symbol;/> ,/> is the gravitational acceleration experienced by land vehicles,/> is the geocentric radius, L _b is the geodetic latitude;/> is the 3-dimensional identity matrix,/> is a three-dimensional 0 matrix,/> is the attitude transfer matrix from the carrier coordinate system to the geocentric fixed coordinate system,/> is the relative strength of the carrier.

(2) Construct a GNSS/SINS tight combination measurement equation.

Based on step 1, obtain the GNSS observation vector (including pseudorange and carrier phase), and make a difference with the observation vector (including the estimated pseudorange and carrier phase) calculated by SINS to obtain the first observation vector; based on the first observation vector , and combine the information assigned by the main filter and the measurement noise covariance matrix obtained in step 2 to update the measurement information.

First, the measurement equation of the GNSS/SINS tight combination can be expressed as:

(8)

Among them, the first observation vector Y _t can be expressed as:

(9)

In the formula, the subscript IF means no ionosphere; m _GNSS means the observation vector of GNSS; and/> are the carrier phase and pseudorange in the GNSS observation vector respectively, that is, the ionosphere-free combined carrier phase and pseudorange measurement values;/> Represents the observation vector calculated by SINS;/> The geometric distance from the satellite to the receiver calculated for SINS;/> is the sum of error corrections related to the carrier phase and the receiver clock;/> is the sum of error corrections related to pseudorange and receiver clock; M _w is the wet mapping function;/> For the zenith wet delay;/> is the carrier wavelength;/> is the carrier phase ambiguity;/> is the sum of other error correction terms of the ionospheric-free combined carrier phase;/> is the sum of other error correction terms of the pseudorange.

Secondly, the measurement relationship matrix H _t is expressed as follows:

(10)

(11)

(12)

(13)

(14)

In the formula, is the ionospheric-free combined sight range Jacobian matrix;/> is the transformation matrix of the position disturbance error from the navigation coordinate system to ECEF;/> It can be expressed as:

(15)

For measurement noise, it is expressed as:

(16)

In the formula, and/> are GNSS positioning and speed measurement errors respectively.

(3) Construct a 5G/SINS tight combination measurement equation.

Based on step 1, obtain the observation vector of the 5G base station (including the azimuth angle, altitude angle and signal propagation time), and make the difference with the observation vector calculated by SINS (including the calculated azimuth angle, altitude angle and signal propagation time), and get The second observation vector; based on the second observation vector and combined with the information allocated by the main filter and the measurement noise covariance matrix obtained in step 2, the measurement information is updated.

The measurement equation of the 5G/SINS tight combination can be expressed as:

(17)

In the formula, is the second observation vector;/> ,/> ,/> are respectively the TOA, azimuth angle (Azimuth, Azi), and elevation angle (Elevation, Ele) observed by the i -th base station at time t ;/> ,/> ,/> are the TOA, Azi, and Ele of the i - th base station calculated by SINS at time t ;/> is the measurement relationship matrix, expressed as:

(18)

in,

(19)

in, It is the measurement noise vector, including the measurement noise v of TOA, the measurement noise v _Azi and v _Ele of _TOA and AOA, and AOA includes azimuth angle (Azimuth, Azi) and elevation angle (Elevation, Ele), /> It is a 0 matrix with 1 row and 3 columns.

(4) GNSS/SINS sub-filter and 5G/SINS sub-filter updates.

Discretizing the state equation shown in Equation (6) and the measurement equations shown in Equation (8) and (17), the time update process of each sub-filter is:

(20)

The measurement and update process of each sub-filter independently is:

(twenty one)

In the formula, the subscripts i and k represent the corresponding sub-filters and discretization time respectively; P _i,k is the state error covariance matrix of the i-th sub-filter at time k ; Q _i,k is the i-th sub-filter The process noise covariance matrix of the filter at time k ; R _i,k is the measurement noise covariance matrix of the i-th sub-filter at time k ; K _i,k is the Kalman gain matrix of the i -th sub-filter at time k ; H _i,k+1 is the measurement relationship matrix of the i-th sub-filter at time k +1; Y _i,k+1 is the observation vector of the i -th sub-filter at time k +1 (the first observation vector or second observation vector); F _i,k+1 is the state transition matrix of the i -th sub-filter at time k +1; It is the state vector at time k predicted by the i- th sub-filter (ie, the local optimal solution).

Step 5. Main filter information fusion and information distribution. The updated measurement information ( P _i,k , Q _i,k and ), perform information fusion in the main filter to obtain the final positioning result, and combine the fused information (/> , P _g,k and Q _g,k ) are assigned to the first sub-filter and the second sub-filter.

After fusing the local optimal solutions of each sub-filter, the main filter information fusion equation is obtained:

(twenty two)

Only time updates are performed in the main filter, and according to certain information distribution criteria, the global optimal estimate in equation (21) , the state error covariance matrix P _g,k and the noise covariance matrix Q _g,k are fed back to each sub-filter. The information distribution criterion when fed back to each sub-filter can be expressed as:

(twenty three)

In the formula, β represents the information distribution coefficient, its value is related to the overall performance of the federated filter, and satisfies the following information conservation principle:

(twenty four)

The size of β is dynamically adjusted according to the PDOP value and carrier-to-noise ratio of GNSS to improve the fault isolation capability of the sub-filter. The global filtering accuracy after fusion can also be improved. β can be expressed as:

(25)

Among them, the PDOP value is the spatial geometric intensity factor of satellite distribution. Generally, when the satellite distribution is better, the PDOP value is smaller, and generally less than 3 is an ideal state; β _PDOP is a constructor related to the PDOP value, expressed as:

(26)

In _the formula _, X ₁ and _{_} ; On _{the contrary} , when the PDOP value is greater than the threshold

(27)

In the formula, Y ₁ and Y ₂ are the carrier-to-noise ratio thresholds, which are 33dBHz and 50dBHz respectively; β _CN is a monotonically increasing function, indicating that the larger the carrier-to-noise ratio CN , the higher the credibility of the GNSS measurement information.

Step 6. Output the final positioning result.

The output includes the fused state vector (i.e., final positioning result, global optimal estimate) and state error covariance matrix/> . At the same time, the global optimal estimate/> Return SINS as a priori information for the mechanical orchestration equations of the next epoch.

This embodiment provides an elastic fusion positioning method in indoor and outdoor seamless environments. It realizes tight combination of observation values in the sub-filter. At the same time, a fault detection and processing module is designed to suppress the influence of abnormal observation values. Finally, in the main filter The processor dynamically fuses and allocates information based on the quality of the observations. Through comparative analysis of simulation and measured data, the results show that this embodiment can effectively isolate sensor faults and non-line-of-sight errors. In the indoor and outdoor transition areas and under the condition that the satellite is partially out of lock, the positioning accuracy is improved by 68% compared with the GNSS-SINS tight combination. %, the effective range is larger than the integrated navigation algorithm based on GNSS, ultra-wideband and SINS, it can work continuously and stably in indoor and outdoor seamless environments, and the positioning accuracy is higher.

Embodiment 2

The purpose of the second embodiment is to provide a flexible fusion positioning system in indoor and outdoor seamless environments.

A data acquisition module configured to: acquire GNSS observation vectors, base station observation vectors, and SINS-derived observation vectors;

A fault detection module configured to: update the measurement noise covariance matrix based on the GNSS observation vector and the base station observation vector;

The measurement information update module is configured to: input the GNSS observation vector and the SINS estimated observation vector into the first sub-filter for difference, and combine the measurement noise covariance matrix and the information allocated by the main filter to perform measurement The measurement information is updated; at the same time, the base station observation vector and the SINS-derived observation vector are input into the second sub-filter for difference, and the measurement information is updated based on the measurement noise covariance matrix and the information allocated by the main filter;

A positioning module configured to: fuse the updated measurement information of the first sub-filter and the second sub-filter in the main filter to obtain the final positioning result, and distribute the fused information according to the information distribution criteria. to the first sub-filter and the second sub-filter.

It should be noted here that each module in this embodiment corresponds to each step in Embodiment 1, and the specific implementation process is the same, which will not be described again here.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Although the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, they do not limit the scope of the present invention. Those skilled in the art should understand that based on the technical solutions of the present invention, those skilled in the art do not need to perform creative work. Various modifications or variations that can be made are still within the protection scope of the present invention.

Claims (10)

1. An elastic fusion positioning method in indoor and outdoor seamless environments, which is characterized by including: Obtain the GNSS observation vector, base station observation vector and SINS calculated observation vector; Based on the GNSS observation vector and the base station observation vector, update the measurement noise covariance matrix; The GNSS observation vector and the SINS-derived observation vector are input into the first sub-filter for difference, and the measurement information is updated based on the measurement noise covariance matrix and the information allocated by the main filter; at the same time, the base station observation vector is combined with the The observation vector calculated by SINS is input into the second sub-filter for difference, and combined with the information of the measurement noise covariance matrix and main filter allocation, the measurement information is updated; The updated measurement information of the first sub-filter and the second sub-filter is fused in the main filter to obtain the final positioning result, and the fused information is distributed to the first sub-filter and the second sub-filter according to the information allocation criteria. Two sub-filters. 2. A flexible fusion positioning method in indoor and outdoor seamless environments as claimed in claim 1, characterized in that the GNSS observation vector includes pseudo-range and carrier phase. 3. A flexible fusion positioning method in indoor and outdoor seamless environments according to claim 1, wherein the base station observation vector includes the azimuth angle, altitude angle and signal propagation time of the terminal relative to the base station. 4. A flexible fusion positioning method in indoor and outdoor seamless environments as claimed in claim 1, wherein the step of updating the measurement noise covariance matrix is: Obtain the observation innovation vector, combine it with the judgment threshold, and calculate the variance expansion factor; After vector diagonalizing the variance expansion factor, the measurement noise covariance matrix is updated. 5. A flexible fusion positioning method in indoor and outdoor seamless environments as claimed in claim 1, characterized in that the state error covariance matrix in the fusion information is a state error covariance matrix obtained by two sub-filters. the reciprocal of the sum; Alternatively, the noise covariance matrix in the fusion information is the reciprocal of the sum of the process noise covariance matrices obtained by the two sub-filters. 6. A flexible fusion positioning method in indoor and outdoor seamless environments as claimed in claim 1, characterized in that the global optimal estimate in the fusion information and the state error covariance matrix obtained by the two sub-filters are , the local optimal solution obtained by the two sub-filters is related to the state error covariance matrix in the fusion information. 7. A flexible fusion positioning method in an indoor and outdoor seamless environment as claimed in claim 1, characterized in that the information distribution coefficient in the information distribution criterion is based on the spatial geometric intensity factor and carrier distribution of GNSS satellites. Adjust the noise ratio. 8. An elastic fusion positioning system in indoor and outdoor seamless environments, which is characterized by including: A data acquisition module configured to: acquire GNSS observation vectors, base station observation vectors, and SINS-derived observation vectors; A fault detection module configured to: update the measurement noise covariance matrix based on the GNSS observation vector and the base station observation vector; The measurement information update module is configured to: input the GNSS observation vector and the SINS estimated observation vector into the first sub-filter for difference, and combine the measurement noise covariance matrix and the information allocated by the main filter to perform measurement The measurement information is updated; at the same time, the base station observation vector and the SINS-derived observation vector are input into the second sub-filter for difference, and the measurement information is updated based on the measurement noise covariance matrix and the information allocated by the main filter; A positioning module configured to: fuse the updated measurement information of the first sub-filter and the second sub-filter in the main filter to obtain the final positioning result, and distribute the fused information according to the information distribution criteria. to the first sub-filter and the second sub-filter. 9. A flexible fusion positioning system in indoor and outdoor seamless environments as claimed in claim 8, wherein the GNSS observation vector includes pseudo-range and carrier phase. 10. A flexible fusion positioning system in an indoor and outdoor seamless environment according to claim 8, wherein the base station observation vector includes the azimuth angle, altitude angle and signal propagation time of the terminal relative to the base station.